Kant and the Demands of Self-Consciousness (review)

Pierre Keller finds an argument for transcendental idealism in Kant's conception of the self. Keller argues that, for self-consciousness to be possible, there must be not only an empirical distinction between representations in inner sense and objects outside us in space but also a transcendental distinction between an ontological inner and outer: "We distinguish the inner from the outer within experience, and the outer as what is completely independent of the way we must experience the world, on the basis of the consciousness that we have of ourselves" (236).

This ultimate conclusion is reached through an argument begun with a suggestive analysis of Kant's theory of self-consciousness. Keller agrees with many other recent commentators that Kant's theory of mind is functionalist. But to this analysis Keller adds an impersonal account of transcendental self-consciousness. This move can be understood as follows: Knowledge of the self as a particular identifiable empirical individual person depends upon objective evidence of the continuity of self accessible from a third person point of view. Objective knowledge is required because mere inner representation of the self as identical over time might err. This objective knowledge, however, is possible only if there is an empirical self-consciousness or awareness of self in inner sense in which different representations are connected together to form a point of view on that objective realm; thus empirical self-consciousness is the basis for but not the same as personal identity. Empirical self-consciousness, the existence of a particular point of view, in turn depends upon the more abstract transcendental self-consciousness which is the transcendental condition for the existence of particular points of view as grounded in the totality of all possible points of view (72). The transcendental "I think" which Kant holds to be required for knowledge, then, is a single collection of all possible points of view, or, a possible grand function synthesizing all possible representations to generate an objective world. Keller holds each particular "I," then, to be a subset of this possible (not actual!) global self-consciousness.

Here Keller's argument is unconvincing. Kant need not hold that particular points of view in inner sense require a conception of all possible points of view in a global whole of representations. Rather, in order to guarantee the objectivity required for a point of view in inner sense, Kant could hold that there must be objective relations among objects which account for the particular point of view even if those objective relations are not representable to any possible point of view. Keller's overlooking this option is puzzling given the emphasis on objectivity he develops in the remainder of the book.

Keller's interpretation of self-consciousness ultimately leads to transcendental idealism through the requirement of objectivity. For since each individual self-consciousness can claim to be such only in reference to a possible global self-consciousness, that global self-consciousness provides the objective background for all subjectivity claims. I can become aware of my place in the world only by becoming aware of other possible relations to the objective world which I might have had. The Refutation of Idealism ultimately makes this claim regarding the necessity of the existence of real objects outside of me in space as the basis for my ability to have any subjective relations at all. Although some have taken Kant's argument here to also refute his own transcendental idealism, Keller holds that this empirical realism can be taken to support transcendental idealism. We must not only have knowledge of objects which we represent as outside of us, such knowledge also cannot ultimately arise from active, spontaneous self-affection, for then the represented object would not be truly able to provide an independent marker for our subjective relations. Thus, there must be objects to which we relate passively; moreover, this passive relation dictates that the objects are "radically independent of us" (216). Hence, the refutation of idealism proves not only that there must...